Jenkins build is back to normal : SystemML-DailyTest #1105

See

Re: Performance test issues

Hello Arvind, Our performance tests have not been tested with HDFS. I am currently working on adding this support. Regards, Krishna On Tue, Jul 11, 2017 at 3:46 PM, Krishna Kalyan wrote: > Hello Arvind, > Thanks for testing these scripts. I will look into this as soon as > possible. > > Regards

Re: Performance test issues

Hello Arvind, Thanks for testing these scripts. I will look into this as soon as possible. Regards, Krishna On Mon, Jul 10, 2017 at 2:59 PM, Arvind Surve wrote: > Hi Krishna, > I am trying to run perftest using following command on Spark cluster. I am > wondering if this was ever tried on HDFS

Re: Decaying performance of SystemML

Thank you Matthias! I'll follow your suggestions. Regarding TB, I had this confusion that "g" implies 512 mb. That's why I kept around 2TB memory. Thanks again! Arijit From: Matthias Boehm Sent: Tuesday, July 11, 2017 10:42:58 PM To: dev@systemml.apache.org Su

Re: Decaying performance of SystemML

without any specifics of scripts or datasets, it's unfortunately, hard if not impossible to help you here. However, note that the memory configuration seems wrong. Why would you configure the driver and executors with 2TB if you only have 256GB per node. Maybe you observe an issue of swapping.

Decaying performance of SystemML

Hi, I'm creating a process using systemML. But after certain period of time, the performance decreases. 1) This warning message: WARN TaskSetManager: Stage 25254 contains a task of very large size (3954 KB). The maximum recommended task size is 100 KB. 2) For Spark, we are implementing this

Build failed in Jenkins: SystemML-DailyTest #1104

See -- [...truncated 28696 lines...] 17/07/11 04:00:00 INFO scheduler.DAGScheduler: running: Set(ShuffleMapStage 1) 17/07/11 04:00:00 INFO scheduler.DAGScheduler: waiting: Set(ResultStage 2)